Study print media rhetoric relating to the Syrian refugee crisis. Run one of the _ex.py files to get started.

• Corpus Parser: Saves JSON articles in a specified directory as Strings.

• Unigram Stats: Counts the number of appearances of each word in a corpus.

  • unigram_stats_ex.py path/to/archive/ path/to/stop-word-list

• N-gram Stats: Counts the number of appearances of any phrase in a corpus.

  • ngram_stats_ex.py path/to/archive/ path/to/stop-word-list (n-gram length)

• Proximate Unigrams: Lists the words near a given word in a corpus.

  • proximate_words_ex.py path/to/archive/ path/to/stop-word-list (offset distance) (word to look for)

• Proximate N-grams: List the phrases near a given word in a corpus in order of frequency.

  • proximate_ngrams_ex.py path/to/archive/ path/to/stop-word-list (n-gram length) (word to look for) (offset)

• Naive Sentiment:

  • naive_sentiment_ex.py path/to/archive/ path/to/stop-word-list (n-gram length) (word to look for) (offset)

• Aggregate Attributes: Lists the number of words with a given attribute in the corpus in order of frequency.

  • attribute_agg_ex.py path/to/archive/ path/to/stop-word-list

• Attribute Dictionary: For each attribute, lists the number of words in a corpus with that attribute.

  • attribute_agg_ex.py path/to/archive/ path/to/stop-word-list attribute

A few notes:

• There must be files ./lexicons/positive-words.txt and ./lexicons/negative-words.txt to run aive Sentiment
• A stop word list must be present. Many such lists are available online.
• The Harvard Inquirer Excel file must be saved as a .csv and be in ./lexicons/inquirerbasic.csv